Bone marrow transplantation (BMT) is a therapy that is performed with curative intent for patients with acute leukemia. The most significant pathogen limiting the overall success of BMT is human cytomegalovirus (HCMV). Previous attempts to control or prevent HCMV infection have been unsuccessful. The approach developed in this proposal is based on the transfer of HCMV specific immunity from the donor to the recipient and on the demonstration that protein vaccines can elicit immune responses and can protect against development of disease. HCMV proteins for potential use as vaccines will be prepared in large quantities using recombinant DNA technology. Vaccines consisting of one or more purified HCMV proteins will be tested in the presence or absence of adjuvant in 20 normal adult volunteers, 10 seropositive and 10 seronegative. Volunteers will be injected 2-3 times at monthly intervals with 50 mug of the test protein, and blood samples, collected at specified intervals, will be tested for in vitro immune reactivities indicative of augmented HCMV responses. Protein vaccines which produce satisfactory results in normal volunteers will then be used to immunize bone marrow donors prior to transplantation. Injection schedules will be based on results in normals. Blood samples collected at specified intervals and bone marrow collected at the time of transplantation will be examined in vitro for evidence of HCMV- specific immune reactivities. Serial blood samples will also be obtained and likewise examined from recipients of bone marrow from immunized donors. The results obtained should provide information about the safety and utility of protein vaccines for HCMV and the transfer of boosted immunity by BMT. This has broad implication for modulation of BMT recipient immunity by manipulation of the donor prior to transplantation, as well as for utility of protein vaccines in other groups of patients susceptible to severe HCMV-related disease.